We have continued to investigate Gammagard lots and their recipients' sera and to determine the reasons for the infectivity of the product. With a nested RT-PCR method (PCR1) which utilizes a two-enzyme system (reverse transcriptase and DNA Taq polymerase), we found HCV RNA in 25 of 43 lots prepared greater than 99% from anti-HCV EIA-2 screened plasma with titers ranging from 2 to 3000 PCR units/g IgG. A similar nested RT-PCR method (PCR2) which employs an one-enzyme system (rTth DNA polymerase) was recently developed and used due to its greater sensitivity, i.e., about 4-fold more than PCR1. With PCR2, HCV RNA was detected in 32 of the 43 lots ranging from 10 to 7000 U/g IgG. In contrast, by either method, only one (2 U/g IgG) of 13 lots from anti-HCV unscreened plasma and one (22 U/g IgG) of 9 lots from screened plasma, i.e., greater than 4 or about 100% screened by EIA-1, were HCV RNA positive. HCV genotyping analysis revealed that some lots had single genotypes but many lots had mixed types. HCV RNA was detected in 80% of serum specimens provided by the CDC. Some have been genotyped; a majority were a single type but some were mixed types. The most prominent genotype in Gammagard lots and recipients' sera was 1a. The buoyant density of HCV in a Gammagard lot having the highest titer of HCV RNA and associated with the most case reports was 1.08 g/mL while that present in an intramuscular immune globulin prepared from anti-HCV unscreened plasma was 1.16 g/mL. The low buoyant density found in Gammagard suggests the presence of free virion (rather than immune complexed virion with low infectivity), which might have rendered the product infectious. Immunoglobulin products which have no viral clearance procedures in their manufacture have been subjected to the lot-release testing of HCV RNA since December 27, 1994. Because PCR2 provided a greater sensitivity in detecting HCV RNA in our low-titer HCV-positive reference standard (lot 2, a positive control for every assay), it has been used since March 1,1996, for lot-release testing. A total of about 130 lots of intramuscular immunoglobulins (IGIM) from 7 manufacturers were assayed. Few lots have been found HCV RNA positive, mainly because all manufacturers have been provided with PCR protocols along with lot 2 and have pretested their lots before submission. Possible transmission of hepatitis C by IGIM products has been suggested in several MedWatch reports. To follow up, we tested the lots and, if possible, the recipients' sera. One case which involved a use of 2 mL of an IGIM product for hepatitis A prophylaxis has been intensively investigated. The implicated lot contained a very low-level of HCV RNA which was only detected by PCR2 but not PCR1. There was no serum prior to the injection available. The patient's serum specimen obtained about two years post injection and approximately at the end of 6-month interferon therapy was positive for anti-HCV and HCV RNA and had elevated alanine aminotransferase (ALT). Both the lot and the serum was genotyped as 1a. Sequencing analyses from NS3 and NS5b regions are ongoing. Phylogenetic analyses will be performed. A pilot evaluation of the risk of HCV transmission associated with recipients of the same IGIM lot in one clinic will soon be initiated by the CDC and such epidemiologic follow-up should provide further insight. confirmation insight. confirmation.